ation, we reconstituted the reaction from purified buy RS 1 bovine heart mitochondria. It has long been established that mitochondria isolated from heart tissue are highly pure, and are functionally robust within cell-free assays. We first tested whether mitochondria isolated from heart were able to generate MDVs by embedding the glutaraldehyde-fixed mitochondria within low melt agarose prior embedding and sectioning for ultrastructural analysis. Indeed, heart mitochondria revealed the presence of 60100 nm vesicular structures still attached to a few mitochondria. In most cases, the presence of both membranes is clearly visible, often with a constriction site evident. We also observed the outer membrane forming protrusions with constrictions at their base, consistent with the formation of single membrane bound vesicles. The presence of these structures suggested that bovine heart mitochondria were an appropriate source for the development of an in vitro reconstitution assay. We developed the cell-free budding assay based on established systems used in the earlier ER budding assay systems. Purified mitochondria were incubated for 60 min at 37uC in the presence of cytosol and an energy regenerating mixture. The buffer systems, concentrations of mitochondria used, and time scale of these reactions were established previously in the development of mitochondrial protein import assays. Following the reaction, intact mitochondria 23382385 were separated by centrifugation at 8,0006g for 20 minutes. The post-mitochondrial supernatant was then treated with trypsin to remove any background signal resulting from broken mitochondria. The trypsin-resistant proteins incorporated within MDVs were examined by Western blot. The assay was stimulated by the presence of cytosol, with peak activity at 3 mg/ml cytosol, and was time-dependent. To quantify the relative enrichment of each mitochondrial protein within the supernatant fractions, we compared the signal intensity relative to that found in the initial isolated mitochondrial pellet. The data reveal that VDAC is more enriched in vesicles compared to the complex III subunit Core 2 . The differential enrichment of cargo found within the supernatant fraction provides evidence for cargo selectivity in this in vitro reconstitution assay. Oxyblot Protein oxidation levels were monitored using the oxyblotTM protein oxidation detection kit by 26771351 chemicon International, following manufacturers instructions. Protein concentration of supernatants containing the MDV fraction was determined by Bradford assay. 20 mg of supernatant were solubilized in 6% SDS with 50 mM DTT in a final volume of 100 ml. Samples were then incubated with 10 ml of 1X DNPH solution for 15 minutes at room temperature. The reaction was quenched with 7.5 ml of neutralization solution in 
all tubes. Following neutralization treatment, loading buffer containing 2-mercaptoethanol was added and the samples were separated by SDSPAGE, transferred to nitrocellulose membranes and blotted. Membranes were blocked for 1 h in the blocking/dilution buffer. Membranes were then incubated for 1 h at room temperature with the antiDNP primary antibody diluted at 1:150 in blocking/Dilution buffer just before use. Membranes were washed with PBS-Tween. Membranes were then incubated for 1 h at room temperature with the secondary antibody diluted at 1:300 in blocking/Dilution buffer just before use. Finally, membranes were washed as above before the incubation with ECL. mtDNA extraction